AP1 cluster - Socket connection failures

What to do

• Check whether you use Pusher Channels WebSocket client API, Channels presence channels. Check your stack

  Paste a package.json, lockfile, Terraform or a list of domains at down.guide/stack to see every provider you depend on.

• Confirm your retries, queues and scheduled jobs caught up after the 1h 45m it lasted.

  Anything that failed between 2025-10-27 11:50 UTC and 2025-10-27 13:34 UTC may need a replay.

• Subscribe to Pusher's feed to hear about the next one first. RSS feed

  https://down.guide/feed?provider=pusher works in Slack (/feed subscribe), Discord feed bots and any RSS reader.

Updates from Pusher 5

1. investigatingOct 27, 2025, 11:50 AM UTC

   We are currently experiencing disruptions with socket connections in AP1 cluster, which began at 10:00 UTC. Clients may experience reconnections. Our team is actively investigating the issue and working to restore full connectivity as quickly as possible.

   provider

2. identifiedOct 27, 2025, 12:22 PM UTC

   A fix is implemented. We are seeing improvements and continue to actively monitor the situation to ensure full stability. During the incident, many clients successfully connected using fallback protocols.

   provider

3. monitoringOct 27, 2025, 12:41 PM UTC

   We observed unusual activity on the cluster, likely caused by at least one customer experiencing a cyber attack. Our team is mitigating the issue while simultaneously increasing cluster resources to minimize impact on other customers. The cluster availability is returning to normal. We continue to monitor closely.

   provider

4. monitoringOct 27, 2025, 12:53 PM UTC

   All services are now fully operational. We continue to monitor the cluster.

   provider

5. resolvedOct 27, 2025, 1:34 PM UTC

   This incident has been resolved.

   provider

Corrections

None so far.

Something wrong on this page? Report an error or send us a note. Every fix is listed here.